Can we really know who’s going to win? Election polling and what it means

Setting political views aside, it is safe to say that Donald Trump was an unexpected candidate in the 2016 election, and perhaps an even more unexpected winner. The majority of polls across the country had him behind Hillary Clinton throughout the 2016 election season. Aren’t polls supposed to be accurate and serve as an indication of how an election is going to turn out? This phenomenon has caused many people to be skeptical of the poll results a little less than two months before the November election. The purpose of this article is to educate on the technicalities of polling, how numbers and percentages are determined and what populations they are representative of in an effort to get down to the true purpose that polls serve in electoral politics.

A publication from the Pew Research Center claims that the purpose of election polls is to “help journalists and citizens understand the meaning of the campaign and the election.” Issue preference, engagement in elections, opinions about candidates, voter preferences and campaign opinions are the main topics of polls. As of September 2, a poll from CNN has Joe Biden leading against Donald Trump, with Biden at 51% and Trump at 43%. Where do these numbers come from? How are the samples chosen? Typically these polls are conducted through telephone calls to a variety of registered voters and people of voting age. Polling typically occurs in the months leading up to an election and as voters are leaving voting booths (referred to as exit polls). Primary elections are a bit more difficult to predict since there are multiple candidates that serve to represent a variety of agendas. Early voting and absentee voting additionally create new polling dynamics. 

An article from the POLITICO website written by Steven Shepard looks specifically at exit polls, with their purpose being to guide the media as national elections are happening in real time. As stated at the beginning of this article, there has been widespread confusion and inaccurate data throughout multiple presidential elections, the 2016 election and the 2004 election between John Kerry and George W. Bush being prime examples. Shepard indicates that the reason for this is that polls have a tendency to have biases and an overrepresentation of young and educated voters. Focusing on these populations tends to create a skew in favor of Democratic candidates. It is important to note that media sources get their polling results from different third-party research organizations, and this can have an impact on results as well. Fox News gets its information from AP VoteCast, whereas CNN, CBS, ABC and NBC get theirs from Edison. Since the results of the polls are coming from different places, there is a likelihood that the results will be different. Shepard states that “the trajectory of election night could look different based on where Americans get their news.” This is important to note, and probably contributes to the confusion that many Americans have when they hear about the different election polls being taken. Over the next couple of weeks, students should be sure to gather news about polling from different sources in order to help mitigate this phenomenon. It will be interesting to see if the polls line up with the election this November, and hopefully pollsters learned some valuable lessons from the 2016 election.